MicroRNAs are some of the most conserved elements present in genomes and have crucial gene-regulatory roles that impact both development and disease. Because of their conservation they have been put forward as candidate phylogenetic markers and, importantly, have the potential to assess completeness of draft-genome assemblies. With the advent of novel sequencing methods thousands of new genomes are now available, but the majority lack microRNA annotations. This large discrepancy is due to the fact that the prediction of microRNA complements typically requires expression data, and no reliable in silico method currently exists.
Taking advantage of manually curated datasets and machine learning models, we developed MirMachine, a miRNA prediction tool. Now we want to develop a web application for MirMachine. We also want to improve its capacity in terms of speed and accuracy.
The development will be done at IFI, UiO, and at the Cancer Registry of Norway, situated at the Radium Hospital. The candidate will learn about bioinformatics, machine learning, statistics, RNAs, web application development in biological contexts. The candidate is expected to learn/have skills in Python programming, R scripting, Linux shell and web application programming.
Supervisors: Dr. Sinan U. Umu and Dr. Trine B. Rounge